Method and Device for Detection and Tracking of Low-Texture Objects

1. A method comprising: at a device with one or more processors, non-transitory memory, and an image sensor: obtaining image data that corresponds to a physical perspective of a physical object relative to the device, wherein the image data includes a grid arranged on the physical object; synthesizing, from the image data, a surface mask characterizing a portion of a surface of the physical object based on the grid arranged on the physical object; and generating a virtual perspective of a virtual model, associated with the physical object, having a virtual surface oriented based on the surface mask of the physical object according to a determination that the virtual surface matches the surface mask of the physical object within a predefined confidence threshold.

2. The method of claim 1 further comprising, displaying, on a display of the device, the virtual perspective of the virtual model associated with the physical object and concurrently displaying augmented reality (AR) content overlaid on the virtual perspective of the virtual model associated with the physical object.

3. The method of claim 1 , wherein the grid lines are at least one of painted on the object, etched into the physical object, projected onto the physical object, stitched into the physical object, or adhered onto the physical object.

4. The method of claim 1 , wherein the physical object corresponds to a physical architectural model.

5. The method of claim 1 , wherein the physical object corresponds to a physical model of a landscape or terrain.

6. The method of claim 1 , wherein synthesizing the surface mask characterizing the portion of the surface of the physical object includes: determining, from the image data that corresponds to the physical perspective of the physical object relative to the device, a plurality of grid intersection points associated with the surface of the physical object having the grid arranged thereon; determining a respective relative height for each of the plurality of grid intersection points associated with the surface of the physical object; and determining a position of the device relative to the surface of the physical object.

7. The method of claim 6 , wherein synthesizing the surface mask characterizing the portion of the surface of the physical object further includes: selecting one of the plurality of grid intersection points as a local origin; and determining coordinates for other grid intersection points among the plurality of grid intersection points relative to the local origin.

8. The method of claim 1 further comprising, after generating the virtual perspective of the virtual model associated with the physical object, updating a location of the device relative to the physical object as the device moves.

9. The method of claim 1 further comprising capturing, via the image sensor, the image data that includes the physical perspective of the physical object.

10. A non-transitory memory storing one or more programs, which, when executed by one or more processors of a device with one or more image sensors and one or more displays, cause the device to: obtain image data that corresponds to a physical perspective of a physical object relative to the device, wherein the image data includes a grid arranged on the physical object; synthesize, from the image data, a surface mask characterizing a portion of a surface of the physical object based on the grid arranged on the physical object; and generate a virtual perspective of a virtual model, associated with the physical object, having a virtual surface oriented based on the surface mask of the physical object according to a determination that the virtual surface matches the surface mask of the physical object within a predefined confidence threshold.

11. The non-transitory memory of claim 10 , wherein the programs further cause the device to, display, on the one or more displays, the virtual perspective of the virtual model associated with the physical object and concurrently display augmented reality (AR) content overlaid on the virtual perspective of the virtual model associated with the physical object.

12. The non-transitory memory of claim 10 , wherein the grid lines are at least one of painted on the physical object, etched into the physical object, projected onto the physical object, stitched into the physical object, or adhered onto the physical object.

13. The non-transitory memory of claim 10 , wherein the programs cause the device to synthesize the surface mask characterizing the portion of the surface of the physical object by: determining, from the image data that corresponds to the physical perspective of the physical object relative to the device, a plurality of grid intersection points associated with the surface of the physical object having the grid arranged thereon; determining a respective relative height for each of the plurality of grid intersection points associated with the surface of the physical object; and determining a position of the device relative to the surface of the physical object.

14. The non-transitory memory of claim 10 , wherein the programs further cause the device to, after generating the virtual perspective of the virtual model associated with the physical object, update a location of the device relative to the physical object as the device moves.

15. A device comprising: one or more processors; a non-transitory memory; one or more image sensors; one or more displays; and one or more programs stored in the non-transitory memory, which, when executed by the one or more processors, cause the device to: obtain image data that corresponds to a physical perspective of a physical object relative to the device, wherein the image data includes a grid arranged on the physical object; synthesize, from the image data, a surface mask characterizing a portion of a surface of the physical object based on the grid arranged on the physical object; and generate a virtual perspective of a virtual model associated with the physical object, having a virtual surface oriented based on the surface mask of the physical object according to a determination that the virtual surface matches the surface mask of the physical object within a predefined confidence threshold.

16. The device of claim 15 , wherein the programs further cause the device to, display, on the one or more displays, the virtual perspective of the virtual model associated with the physical object and concurrently display augmented reality (AR) content overlaid on the virtual perspective of the virtual model associated with the physical object.

17. The device of claim 15 , wherein the grid lines are at least one of painted on the physical object, etched into the physical object, projected onto the physical object, stitched into the physical object, or adhered onto the physical object.

18. The device of claim 15 , wherein the programs cause the device to synthesize the surface mask characterizing the portion of the surface of the physical object by: determining, from the image data that corresponds to the physical perspective of the physical object relative to the device, a plurality of grid intersection points associated with the surface of the physical object having the grid arranged thereon; determining a respective relative height for each of the plurality of grid intersection points associated with the surface of the physical object; and determining a position of the device relative to the surface of the physical object.

19. The device of claim 15 , wherein the programs further cause the device to, after generating the virtual perspective of the virtual model associated with the physical object, update a location of the device relative to the physical object as the device moves.

20. The method of claim 1 , further comprising: forgoing generation of the virtual perspective of the virtual model associated with the physical object according to a determination that the virtual surface does not match the surface mask of the physical object within the predefined confidence threshold.

21. The method of claim 1 , further comprising: identifying the virtual model from a library that includes a plurality of virtual models based at least in part on the image data that corresponds to the physical perspective of the physical object relative to the device.